tb2c.

_____

TECAINICAL MEMORANDUM 1998

-AN ENGINEERING STUDY

..OTPARIN( INSUrLATOW-RESISTANGE
TO DIELECTRIC STRENGTH TESTING
RICHARD J. MEMICE SAFEGUARD QUALITY ASSURANCE DIVISO
-MAY

1971

TIS 17& ', JMENT HAS BEEN APPROVED F'OR PUBULC Y~Lf-MAr AND SALJE; MT DISTRBUTION IS UP4MMMD,

NATIONAL TECHNICAL INFORMATION SERVICE

Springtiold. Va. 22151

Reproduced by

.4?ICATINNYAR E L D3rVER, NEW JERSEY'

UNCLASSIFfED
Security Classlflcation

DOCUMENT COf r

R&D.
" -:1 t tpitf IN cM..ihedj n th& cnvjtt ,* X.* RCPOYOIECUTY C LAWIICATrION

I" re.uity cla it 1 *12- i nf titll*, bfty oft .hejt,.t 41%dLnd~xfn# anMIarC. M'umf f S1ORIGINATING ACTIVITY Corporal# sufhar)

U. S.Army,
1. REPORT TITLE

Unclassified Picatinny Arsenal, Dover, NJ 07801

b

An Engineering Study Comparing Ineulation Resistance to Dielectric Strength Testing

4. OESCUIIPTIVi[ NOTES (rype v( repot and i,eluolg .9 41.)

Z.AUTH

R(S) (La~f"onc "rftff 'Im. tnt,;.,)

Memice, Richard J.
S. REPORT DATE 7.. TO'AL t4O. Cw PAG#* 7b No or steps

May 1971
Se. CONTRACT OR GRANT NO. (L PROJRCT 40.

45
9e. W41tkOINAR atCWOR7IVJtA5RtS)

13

PA Techicat Memorandum 1998

AMCMS Code: 4270. 15. 9000.

C.

6
QT* C PPuORvT I;o(Si ,'A:.y oth~tntm;4er* fhat *,ay'ba, amaii~ed

10. AVA ILABILITY/Li WITATIOI $4OTICES1

This document has been approved for public retease and sale- its distribution Is unlimited.
11. SUPPLCMENTARY NOTES

1n. SPONISORING MILITARY ACTIVITY

Picatinny Arsenal, Dover, New J.rsey

I,. AB,,,.rCT

This-tudy inveatigates the differences, simulations, and interrelationships betweer insulation resistance and dielectric strength testing to dettrmine if either can be eliminated to effect economy in testng.

DD IJW 14-73

UNCLASSMFED Security Clasification

.. .5

S-u ;,

ClSittlu

ti!

fii
Stcut

tic,
IA

LINK A

LIK 0

LINK C

II

__ ____ __

__

__

_I
ho. AVAILABILITY/LIMITATION itations (ton dissemination thr

INSTRJCTIONS 1. ORIGINATING ACTIVITY: Enter the name und address of the contractor, eubcotactor, grantee, Department of Defense activity or other organ!zatlon (cotpogate euthor) issuing the report. NOTICES: Enter any !Lmof the rport. other than those b, security clasiflcstlon, using sta d stements suc ss such ar. (1) "Qualified requastras may obtain copiz of this report om DDC." (2) (3) "Foreign announcement and dissemination of this report by DDC Is not rothoried." "U. S. Government agencies may obtain ctoles of this report directly fm DDC. Other qualified DDC uwer shall reqest tivought ll "U. S. m lltf ec"s may obtain Copies of thIs e A-= a Othe qualified usere shal request trough
"Al l
. .

2a. REPORT SECUITY CLASSIFICATION: Enter the over all security classification of the report. Indicate whehat "Restricted Data' Is Included. Marking Is to be in accord. ance wit" srv"opriate aecurity rtguls!ilon. 2b. ORb, .Automatic downgrading it epecfied in DoD DiI rective S t-d Armed Forces Inidustrial Manual, Enter the group nt. r. Alao. whn applicable, show that optional markitgs hsv been used for Group 3 and Group 4 as sutor i

lsed.I

3. REPORT tTLt EntreA the c@Wite report title in till capital lette-s Titles in all ca shold be uclsslfied. If a meanirgful titlec o- be aelectid withmt classifiestion.-show title clasificailon in All c*pitals In pateah is imm edi a tel y f o l l o w i n g t h e u t l e. 4. DESCRIPTIVX 1?T If appropriate. enter the tpe of rport. e&., interim, prowasa, sommat, annal, o fla1l, Give the inclusive dates whn a specifile reporting period is

(4)

( 5)

__ _ . _.. _ _ _

i s. g a _th

_ _ _O _ _ _ _ _

(5)

"All 4lstrbtl tlo treport iontlle. filed Doc useirs si request thowk

QiZ.

covered.
eoIf 5. S. ift:W emporu F.nter Ist s mmefigt autboeg as ow On the is) of natn; s) or AUTHO S)c E mddleL ltLL Itmilitary, show aink aad branch of service. The some of the principal author is en absolute talis requtrenut. 6. -REPORT DATM Enter the deate he repot as day, month. lew; or month. ym Wafor than one dute appeas on the repaft, use dat of publiletio 7. -TOTAL NUISER OF PAGES The total page count should follow norml paginstion piocedures, Le., enter the number of pages conening normatio. 7b. NUM SER OF RE 'V references cited in the reporU & Enter thv total III, the re has been finlsadhot the Office of Technical SrMlcss, fact eeter Cotiwco. for sale to the public, Indlcata.this DWatmilt of "mprice# i lknawn.

IL SPPLZWLNTARY NOTE& Use for additional esxplanstor mite. 12. WOWNUING MILITARY ACTIVITY Efct the name of the dipartmentu_ project office or I try sponsoring(ar ' 1D4 o,)tinreIw-hIanddeo nt. Lclad adese 13. ABSTRACT: Enter an atract giving a brief and factal !ummary of the document indicative of !he report, even though it may also aMer elsewhere in the body of the technical re. f of I port. be additional sp .e is required, a CoMtittutiott sheet shall If attached.

go. CONTRACT Of GRANT NUBEtb Itapproriate, enter I is highly desirable that the abstract of classiffd itthe applicable number @1 the contract or pstnundet which ports be unclassifled. Each paragraph of the abstract shall the repor was written, j end with an indication of the military security classification Enert the appropriate project umber, subproject mtmber, system ustebre, task number, etc. Enter the offlga. ORIGINATOR'S REPORT NUER(5) cisI report number by which the document will be identified and controlled by the originating activity. This number must be unique to this report. i): Ifte a'spo has been P qb. OTHER REPORT NIJM asigned any other report numbers (either by the orlinator or by the sponsor), also enter this number(s). Sb, 1:, & Id PROJECT UMBEMh

military department identificatlon.

Waches

2 (5). of the Information in the paragraPh. ap rented as (T (C), or (Mf. There Is no limitation an the length of the abstract However, the suggested length is from 150 to 22S words. 14. KEY WORDS: Key words are technically meaningful terms or short phrases that characterize a repert and may be used as index entries for cataloring the report. Key womb must be selected so that no security classification is required. Identiers, such as equipmtnt model designation, trade atme, mlitry oJect ce name, geographic location. may be used as key -words but will be followed by an Indication of technical context. The assigrwr,*nt of links, rules. and weights is c optional

s,

Security Classiicatiom

TECHNICAL MEMORANDUM 1998

AN ENGINEERING STUDY COMPARING NSULATION RESISTANCE TO

DIELECTRIC STRENGTH TESTING

BY

RICHARD J. MEMICE
AMCMS CODE NO. 4270.15.9000.6

MAY 1971

This document has been approved for public release and sale; Its distribution is unlimited.

PICATINNY ARSENAL
DOVEr NEW JERSEY

__.1

TABLE OF CONTENTS

PAGE NO.
List of Illustratons Abstract Summary Conclusions Section L Reasons for Insulation Resistance and Dielectric Strength Testing IL Comparison of Alternating Current to Direct Current 11L Theory of Br 18 20 7 3 4 5 6

IV. The Importanie of 60 Hertz Dielectric

Strength-T ating
V. Factors AffeCtnnulatlolx Pesistance and 23

Dielectric teW
References Glossary Formulas

Testing
32 33 36

Appendix Distribution U*st

37 41

LIST OF ILLUSTRATIONS

FIGURE NO. la
lb 2 3

PAGE NO. AC Dielectric Circuit

DC Dielectric Circuit Magnitude of Current Vs Capacitance in An AC Dielectric Circuit Dielectric ~trength VS Pressure and

10
10 14 25

Altitude
4 5 6 Modified Dielectric Circuit Dielectric Phase and Loss Angle Capacitive Power Factor VS Frequency 37 37 40

ABSTRACT

This study investigates the differences, simulations, and interrelationships between insulation resistance and dielectric strength testing to determine if either can be eliminated to effect economy in testing.

SUMMARY

The object of this study is to determine the most economical test method for insulation resistance and dielectric strength testing. It is common practice in some industries to test by elevating the insulation resistance DC testing voltage by some multiplying factor which has been determined as equivalent to the dielectric etrength AC test; and thus eliminate the dielectric strength test altogether, resulting in economies generated by the savingo in testing time, data reduction and analysis, and test hardware. thanlysis wiWlinvestigate'A.e differences, similarities, and interrelationships between insulitin.u resistance and dielectric strength tenting to determime if there is an overlap or equivalent duplication of testing that could be eliminated, re ling in the above mentioned economies. This-study is also extended to include the theoretical analysis of the various types of failures and breakdowns that can occur In insulation resistance rnd dielectric strength testing., .Thls type of knowledge is necessary before anyone can corpatently e#tbish specification test requirements and test methods.

II

CONCLUSIONS

Aircraft equipment testing found dielectric strength testing of 28-volt equipment at 500 volts rrns and 60 hertz not sufficient in detecting flaws in insulation thicker than one rail, therefore this test will not offer more significant data than an insulation resistance test of 500 volts DC and it will not provide quantitative leakage data.

L REASONS FOR INSULATION RESISTANCE AND DIELECTRIC STRENGTH TESTING

Purpose of Insulation Resistance Test-g The primary purpose of insula-ion resis--ance idirect current) testing is to determine if the ie-Ak3ge currents i,- insulator are low ar enough to ensure safe and relabi operatiou with a minimum of mantenance and repair. In_,ilation resistanciestg checks insulation condition with respect to dryness and consami a on with dirt, oil, and chemicals. Low insulation resistance can form unwzaed paths for current which can disrupt and cause failuxe in the compo ent, serve as false signals, or dissipateeiectrical signals. When leakage currents are excessive, insulation deterioration is accelerated because of internal heating or through electrolysis. Cold flow of the insulation may cause the insulation resistance to be lowered to the poin where large leakage currents nay be formed. Insulation resistance testing can assure adequacy of insulation thickness, proper clearance of parts, and lack of mechanical defects which car cause electrecal breakdowc Insulation resistance tests indicate nof only the immediate condition, but can be used to estimate the probable future life of the insulation by obsenving the value of the Leakage current as the voltage is increased, If the leakage current rises linearly to the final value of the tcst voitage, the life expectancy of the tested item can be ex ected to be good. if the leakage current changes exponentially to the final v-due of the .est voltage, the

life expectancy of the item will probabIly bz- shor.

Insulation resistance testing is affected by the temperature, time and rate of application, level of a pled voltage, moisture, contour of the specimen, environment, contamination, aging, and prpvious history of the isilation. Insulation resistance tes;tig is mainly used as a nondestructive test made on high capacitive items, motors, generators, cable runs, components, assemblies, or complsted equipment to dt&nermine whe ther the insulation level is satisfactorily high to assure reliable operation. Insulatio, resistance testing can minimize te possibility of expensive service failures, can show up cases of ibadequate design, and can lead to a more
effictent product

-i

Purpose of Dielectric Strength Testing The primary purpose of dielectric strength (alternating current) testing is to determine if the dielectric can operate safely ai.d reliably at its rated voltage and if it can withstrad high voltage surges and ti'ansients. Switching transients and high operating voltage have emphasized the need for high quality dielectrics and dielectric strength te~ts to prevent breakdown. Transients caused by apacitive dischargeq are infrequent because they must be charged above line voltage. Trarsients caused by inductive discharges are very common ant usually occur in motors, solenoids, relays, or with any interrupoon inany inductive circuit. The maximumr instantaneous voltage will be equal to *LC I volts where L C I = inductance = capacitance =current

A circuit with a 10 mh inductor, a 100 pf capacitor and a current of 100 ma can have a transient of 1,000 volts. Dielectric strength testing can determine which materials are extremely susceptible to corona damage, carbon~zation, and puncture by high voltage surges. Insulation should withstand dielectric strength testing without rupture, in addition to preventing excessive current flow between two circuits. Solid electrical insulation materials are generally nonhomogeneous and may contain dielectric defects of various kinds. Weak spots within the material usually determine if the test results will be good or bad. Dielectric strength testing is affected by temperature, pressure, time, rate and level of application, moisture or humidity, frequency, thictmess of specimen, electrode configuration, environment, contamination, waveshape, aging and previous test history. Dielectric strength tests can be conducted either with AC or DC, but AC is generally used to determine if large capaciave currents are present in addition to leakage currents. As

a rule, dielectric strength tests are 1_G-PV for production testing of small
ct ?acitive items, for materials testing and anywhere go-no-go information is desired. Defective equipment caught early can prevent extensive and costly repairs and ensure reliability, quality, and safety.
3

-~2

PE_
IL COMPARISON OF ALTERNATING CURRENT TO DIRECT CURRENT The Dielectric Circuit The following differences between insulation resistance and dielectric strength testing were found to exist: a. Dielectric strength tests are usually conducted with alternating current and are mostly used for .- no-go testing of small, low capacitance equipment, for production testing, and materials tests. b. Insulation resistance tests are conducted with direct current and are made-oncomponents, assemblies, or completed equipment items to determine whether the insulation level is high enough to assure reliable operation. They are generally-employed where quantitative, rather than qualitative results,:are neessary or when large equipment or cable runs having-high:-capaitance mustbe tested. c. Because-direct current voltage distribution is inversely proportional to conductivity:and alternating current voltage distribution is determined by permittivity, thermal a.'d-disruptive breakdown will usually occur first with-AC, while Intrinsic breakdown will usually occur first with DC for a given voltage and time. The following similarities between insulation resistance and dielectric strength testing were found to exist: a. As. a dielectric material or component is repetively tested, a lower test voltage will give the same results as did the initial test voltage due to deterioration caused by voltage fatigue. b. Because cdelectric strength factors are not significant at low frequencies if capacitance Is low, a graph (based on capacitive power factor) was generated to-indicate when to use insulation resistance and/or dielectric strength testing. By neglecting the leakage current or the-capacitive current when it is 10 percent or less of the total current, the related test may be omitted.

The following interrelationships between insulation resistance and dielectric strength testing were found tc 2-xist: a. In the early days of the e' -tricai industry, DC high potentials were unobtainable and with the dovelopment of transformers, alternating current generators and motors, most insulation testing was done with A%. Today, with great advances in the field of selenium and silicon qolid-state-reetlfiers, the DC tst set is being used in areas-where AC once predominated. b. There can be no general agreement on an AC rms to DC conversion factor because materials will differ in leakage, and electrolytic action. The quivalent circuit of a dielectric may be represented by a capacitor, C, and resistance Ra and Rb in parallel as shown in Figure 1 (a) when an AC current is used, 0
Eq

Rb

Ia

Ra

Figure 1(a) AC Dielectric Circuit

Figure 1(b) DC Dielectric C rcuit

The capacitance represented by C is between the two metallic circuits, the resistance Ra represents the absorption losses and Rb represents the leakai a paths through and over the insulation. Figure 1 (b) represents the equivalercircuit of a dielectric when DC is used and capacitive and absorption currer,3 are small enough to be neglected.

10

t
Ls

AC Components When an alternating current is applied across the circuit, the total current would consist of an out-of-phase capacitive current, Ta, and an in-ph: Se or leakage current, lb. The AC test set must supply the vector sum of these two currents. When the item under test is low in capacitance, the capacitve current may be small-compared -to the leakage current and may be neglected. When the iteri under test is-high in capacitance, the capacitive current may b many times the leakage current. This statement can be illustrated with the aid of the following example In Figure 1 (a) the impedance of the first branch, Za, can be represented by the following equation: Za = where W C

a + jwC

(a)+

(/C)

ta-

1/wRaC

= 2 ir times the frequency capacitance of item under test

Ra = resistance due to absorption losses The impedance of the second branch, Zb, can be represented by the following equation: Zb =Rb
--

where Rb = resistance due to leakage paths.

The current, I, can be found by dividing the voltage, V, by the total impedan(-!, Z. If we let Ra = 10 7 ohms Rb = 108 ohms w=2 r f = 2ff (60) Currents and impedances can be calculated and are listed in Table 1, page 1". for several values of capacitance with other parameters held cnnst.ant.

11*

TABLE 1 - CURRENT-AND IMPEDENCE FOR SEVERAL VALUES OF CAPACITANCE

cZa C_ 500 10 to50 5001400 10

_Zb L 2 OC6M? 1-9W0 L,66fl0 8 1-880 2.84X61

7 17r

I1

10
.;16O1

i07oi08 L0

0. IRS t:Wrua
1.88-4ts0 17.6 170'

Sa 5 5

108 LO0 O

_5 O* r

L05l 7 ti

48.2 Lr

zoo-tee io-L

-2

0 O 502'-

12

As shown in Figure 2 the magnitude of the capacitive curren Ia, increases as capacitance, increases. When the capacitance is 10- 12 the leakage current is over 25 times the capacitive currr nt and the capaci ive current may be neglected. As cap:icitance increases to l0 - 9 f the capacit -e current is 10 times the leakage curient and in most cases 'h,1 leakage cu: 'ent can be neglected. The ITT Reference Data for Radio Engincers shows thi capacitance per foot for most singl- ard double bradj.d-t cables to be abou, 30 pf (10-12) per foot, but some low capacitance cables can get below 10 pr/ft Depending upon the distance between conducting paths ind the length of the path, the capacitance increases the magnitude of the cnpacitive current w, I also increase. DC Components When a direct potential is applied across the circuit, the tot.i current flow would be the sum of (1) capacitive current, (2) absorption cu rent, and (3) leakage current. The capacitive current decays at an exponential cate according to the equation:

Ic
where

E R
Ic E R t C

exp (-ti/C)
= capacitive current
=

applied voltage internal resistance

time after voltage applicatior capacitance of item under tes'

The absorption current is caused by polarization of electric charges whici. take place in an imperfect dielectric under the stress of an applied voltagi The absorption current also decays exponentially according to the equatior-

13

zz
-

zz
ww

zr
05

w ()

0
-

0
-

to

th
wt

06
v ocl

dh 6
~VTr

0I

NI

INA8fl3 d0 3anflNDVW
14

la vi -,re la A, B V
t

AVct exp (-,

=absorptior

cu,, rent. dep nding upon tb'- mate-rial

constinths

incremen'tal -2ange in voltagstime~ after &i nge

Iin

voltage

capacitance c item under t'r-st

T'leak. e current depends on -the ;avd1-series resistancE of ti circuit (. aris g4- :'1 by the equation, F2 E RB
v, ore:

Jib E

=leakage

curren t

applied voliage Rs total series reststince

B, itude 'e first two compone.nts d !-ay with time, the current measured at 'r suf, l ent time has elapsed for conditions to stabilize is the true Icakage cl -ent.
Aaj A,

of the AC Test Set

W.S c,,. rr

w,
vil

rnating current is prima' iay used in the field of mate-rials t.sting -re hr - -down tests are made on 3amples of irnsulating mlitrrals :).d where 'pes -e small enough for capac Jvp, currents to be nt-g*I':cte d. APl-rnaling '-ent resses the dielectric in pi: portbon ft-bi-de1ectric corstant of the:-. :erialr rather than in proportion to the leakage- resistance- as is the c-Ist, DC 1. sting. Ii the dielectric cc-astent is to he s~ressed or m, c.hnnica1 -ation )roduced by alternatirg ev rrrt, an AC Tes* sef would he required.
'. .

Alternating current test sets are primarily go-no-go testers. As the voltage is raised to a specific level, the item under test may or may not break down. Indication of pass or failure is usually given by a light. The degree to which an item passed or failed is not known. This test will indicate only if the item isi gocd or bad. An AC test set may be inconveniently large when equipment or cables of high capacitance are being tested. The capacitive current could mask abnormally high leakage currents when high capacitance iF resent. In production testing, alternating current permits s the use of a sn.all high-resistance transformer which is safe and easy for the operator to use. Advantages of the DC Te:,t Set A direct current test set can give more than a good or bad indication. It can indicate the degree to which an item passed or failed. Direct current is used when information rego rding the comparative condition of the items tested is needed. The AC tes: set requires a circuit breaker for go-no-go failure indication, while with a DC unit, leakage current is measured and may be compared with a limit"ngvalue.__Catastrophic failure is indicaLed by detection of avalanche breakdown. The acceptable value of leakage current is usually determined by corn arison testing. In a DC test set, the capacitance will have little or no effect on the steady-state value of the leakage current, but a considerable length of time may be needed for the current to decay to the steady-state value. rhe charging currents can be kept within reasonable limits by gradually raising th voitage so that the incremental voltage divided by the series resistance yielh s an initial current within the rating of the test set. If the voltage is raised .-lowly enough, even a highly capacitive item may be tested with a unit capable f delivering only a few milliamps. Direct current test sets are usually aised when the capacitive current of the item to be tested is so high that a very I irge-AG unit would be needed to perform the same test. AC to DC Conversion Factor When direct current is used in place of alternating current, a conversion factor is generally uied. Aercspace industry has been using values between 1. 8 and 2. 0. Many nanufacturers have been using the value of 1. 7. The cable manufacturers hay 3 used 2. 3 and even higher numbers. In the American Institute of Elbc trical Engineers (AIEE), Transactions Paper No. 58-845, the value 1. 414 imes the 60 hz rms sine wave AC test voltage is

16

suggested. The valueoftL7has appeared the most, but itIs not the accepted standard. The reason that there -is no universal factor Is because the dielectric-constant varies- from material- to material and alternating currant stresses a nonhemogeheous material in proportion to Its dielectric c-onstant. Other-factor. Including-different mechanisms for br~eakdown, tr-anspr ofcharges,, su-rfce'leakage effects, electr~olytic action, etc., *"ic-differ from material to maerial iake It virtualy impossible to eati~lh aunveral actrsThe test-valtage to be used in any given -ndthe situation de nds -on the -restults-of-an exp erimental aprch jud~it ofthe tdgt eniner in-general-agreement with the R&D engineer

III.

THEORY OF BREAKDOWN

Thermal Breakdown The thermal theory of breakdown is based on the assumption that all solid dielectrics are heterogeneous. Because of this quality some parts or areas have a lower resistance than other parts of an apparently uniform material. When a current is passed through a sample, it will not.be uniformly distributed. Parts lower in resistance will carry more current and will be heated quicker than parts with lower currents. If the adjacent electrodes or insulation can conduct the beat away as fast as it Is generated, the temperature remains siable and no failure will result. However, if the heat is not removed as rapidly- as it is generated in any part of the dielectric,. parts with higher currents grow hotter, thereby lowe-ng the-resistance still further. As the voltage is -increased, the temperature rises until-thermal. Instability occurs. Therefore, the dielectric will breakdown at its weakest point. ThermalEffect on Insulation Resistance and Dielectric Strength Electrical conductivity rises with increasing itemperature therefore the hottest paf-of-the solid-Insulation-Ii elf~vedof some-of its electrical -strenwith DC where-the:-voltae distributo is-inversely proportional to -. tbe condctivity. TMqrelief is not inhereit-in AC testing-wherethe field distribution is determiled-by the pernitivity of-the material which is usually indepenet oftemprare. Therefbre, an insulation resistance test conducted-under-the:ituleice-of DC mry tolerate-high- ambient temperature where brd may occur ona-dielectric strength test conducted under the

influence.of AC.

--

Iftherma Instabfiity-does not-cause breakdown of-the dielectric, failure may result-when the field intensity becomes sufficiently high-to ac delerate electrons through the material. The-critical- field intensity Is known as the-intrinsic dielectric strength. Ionization will occur from either collision or chemical -action under the-influence of field voltage. As the

-pvolta6e Iancrezsed, ions will dissipate energy and produce other ions. By
increasing-the ,o1tge-a higher field intensity Is formed,' ions are produced at a fastr rate until instabilit- occurs and ins ation failure results.

18

Intrinsic Effect on Insulation Resistance and Dielectric Strength When direct current voltage is increased, electrons leaving the cathode will move toward the anode with greater velocity and fewer of them will return to the cathode. When sdfficient energy is attained, collisions between electrons can free additional electrons from molecules in the process known 98 ionization. With direct current, acceleration of electrons will-be constant and ionization can occur at a maximum rate. Alternating current may cause reversal in the field in less time than ionization canget-started. Therefore, an insulation resistance test using DC will probably cause. intrinsic breakdown before a dielectric strength test using AC for any given voltage. Disruptive or Electric Discharge Breakdown Electrical breakdown is caused by physical- rupture-of the-dielectric resulting in the destruction of molecular and other bondsi This rupture is caused-by electrical charges which are produced by Ngh local fields. fMolid materials are tested, the discharges usually occur in-thesurroUndtng-me rium which increases the test area and produces failure at-orbeyond-th: electrode edges. Discharges usually occur in-internal voids orbubblesthat are-present or may develop. They are caused by local erosion or chemiical decomposition. This process usually continues until a complete failure-path -is formed between electrodes. Discharge vaffect on Insulation Resistance and Dielectric Strength A dischaige across a void is similar to discharging a capacitor. After discharge, the voltage drop across the void itself islowered -and the discharge may stop. Depending on the time constant of=the -material, further discharges may take place with DC. When alternattr4,currovnt isused, the internal discharges can occur during each half cycle Alternating current -will subject the insulation to a vibrating mechanical forgetwhich results -fromthe alternating field and this effect cannot be obtained with steady direct current. Therefore, an AC test will probably cause disruptive breakdown before a DC test.

19

MV THE IMPORTANCE OF 60 HERTZ DIELECTRIC STRENGTH TESTING

t In a report titled "High- Potential Testing of Aircraft Equipment Eectrical Insulation", failureanalysis of 28-volt equipment showed that 500-volt rms, 60-hz, tests were not detecting-insulation flaws which subsqently paused -service-failures. The 2purpose-of- the-E,30-vol't, test Was etet-flas in the-Ansulatidn. Jn operation* inductive .circuit -interruption to tratsients--are ver-y4iy-to:discharge throug -theseflaw8 or._ puntcture m~a 4y -thiinistion-And Ciabe insulation -failurebyztrackig -Circuit transient volteg-discharged- cross voids in the material-were not beingdetected-by this-test.

An-investigation -was made to determine the reason for insulation fdI1ure. The investigation indicated four factors -caused insulatio n-biekkdown. They are:

a. -Voltbge'gradients in-the gap:area-caused-by-the-change 4in dielectile-conustait--from ii-r--to materiaL

b. Limited dissipation-of--heat cautsed-by loss -of-rconvevtioni cooling. c.Contamfinatidodischarge the -afr gaparea-_with -products o h

~~I. Argppesure changes-caused by-confiniaig the-disp

chargs 4n~malldolos
-Underhighly ,controlled--con&diin, tests-were mnade on-materials 1, 5, 3- 7,.a9nd79 -mlithick. R Ultsbowtd-500 -volts-rin, 60 liz, 'tests -would not-detect a flaw- reliably in insulationthicker than one- iiL One thousand-voltsrmswould-proably .be-necded to -d~etcta-fla~w-in mitinsulatin. The-investigation W. als-o-zhowed ta thelarge-the dlaimeter.of-the hole the lower-the breakown -voltager.

20

Govermental Dielectric Withstanding.Voltgge _Tes. The purpose of the Dielectric Withstanding Voltage Test in M11 -STD-202D, Method 301, is "to detertilne whether insulating materials and spacings in the component part are adequate". In governmental testa, Dielectric Withstanding Voltage Tests are usually conducted at 500 volts rms, 60 hz, and according to the previous report on dielectric strergth testing of aircraft equipment, only a one mil flaw can be detected reliably at 500 volts rms and 60 hz. To detect a 3 mi flaw, 1,000 volts rms at 60 hz woild be needed. The aircraft equipment report and various other sources indicate there is little advantage to equipment testing using 500 volts rms AC when 500 volts DC testing is preferred. "Failure analysis of 28-volt equipment repeatedly demonstrated that 500-volt, 60-cycle AC tests were not detecting insulation flaws which subsequently caused service failures. "5 Other sourced can support the airoraft equipment report's findings. American Society for Testing and Materials (ASTI) maintains that "dielectric strength is not significantly influenced by frequency variations in the range of commercial power frequen.cies (50-60 hz). " "The effects of capacitance are discernable at 60 hz. "2 It is a known fact. that an AC current can cause field reversal in less time than avalanche breakdown can get started. History-of High Potential Testing The importance of sound electrical insulation has been recognized from the early days of electricity. The need for improved testing and insulation was emphasized by-damage caused by flashover, lightening, transients, and problems caused by continued use, Most early testing was done with DC potentials of up to-500 volts. Higher DC potentials were not feasible. Later, wfth the development of transformers, AC generators and-motors much insulation of testing turned to AC. As the years passed and the-electrical industry expanded, rotatingmachinery, transformersvnd cables became larger. AC-test equipment became larger-snd more costly to keep pace with the insulation to be tested. The phase-to-phase or phase-to-ground capacitance of some cables became so large that the capacitive current was as high as 2-3 amperes.

21

it was not until the !930's that the invention of the Kenetron reter tube led to the development of direct current high voltage supplies. Because of new designs available with the Kenetron tube, DC voltage supplies were a great improvement when compared -to size, weight, end cost of the AC units. But stil, AC test Bets predominated in the 1930'8 and 1940's because of its longer history and the considerable data that had been accumulated correlating test results. Most testing wa done with AC also because of the lack of general familiarity with the newer, more sophisticated DC test equipment, and the availability of AC equipment, rather than because the AC teat per so is better. As time progressed, tremendus -advances in the electronic in xstry, increasi use of-higb voltages, and the increasing complexity of eqipment, the demand for more sophisticated, more sensitive, and saf&--teat 6uapmtnt increased; The-high reactance transformer made small portAble-units for AC testing fast, convenient to use, and safe to operate. It was not until the 19M0Is

tat the -advent of the high vacuum recters-reduced the size, weigt a4nd Of thi DC set that-DC testiagbecame :popular. Today, with ft great advancs in-the field of selenium and sllicn solid-state rectifiers, -the DC
-ost

test skt is high in quality and -reasonably priced.

In spite of the earzy:hwd .it!.AC-test-sets had, DC test unis are

being usd*-m

here AC p

ftnted. DC test sets are -preferred to

test IftfofMhigh capacitance or when quantitative rather than qualitative

measrmentsmust be made.

-2

V. FACTORS-AFFECTING INSULATION RESISANCE AND) DQIELtGTmC-STRENGTH TESTING

IelecSrht Tbs teinflrature of the teat specimien and its ra' tidltig medium .. influenee the didibtilstrength, but for most materials small variationsofftrabein _--ay haive-a-negig~be effect Dielectric strength.Vwil decrease wihIorsigteprtre, but the extent-to-which this is true ~ ~ I uponthe tst& Raising-the-temperature reduces the ateral nutaber-Lt volumtausing the electrons to travel farther c and ~farther btwacolisions with nitoeculs of the gas, thereby permitting ee a to cause lonintion with every collson, them ~ Up s-fci to pic up

it~th am!the

Imultonresstane i alo afectd b temeraureas dielectric gs oft-differnewl depend on the typa ofinsulation, the

wo*hasaboi tat umuatlc esitac willberedue -by a fixed rato for &_ each1seicteprure increment Most Inltofs have multipliers of temp-erflure, with-the 40 percnt to 75- ercent for ac 100incresin =uMitpir 2cr*eac)kim and cution of insulitln remaining appdroximately costiat over-the operating rng4e.

In,-genera, as pressure Is Increased, -the dlelectric-istrength of the maeialo is I ~ Thlis e -to-the-fact-that apresare 1rs Inceased miOeuleInQthe ga atWare s loser together with the-distance betriet caunien~t~oaw1~~sva oolllo, ibrb nreaslft: the -kreakdow vcl~e.Bne. tis oleIs ot pplcabe t alpresur.As -the gas prawnIs loered ollii ecome less freqent, thuas permittin the eletrns-o ickupsuficentenrgy to aue Ioni-zation and lower the break down voltage. As-the gas presmre Ii-lowered stilt-further, there is a point

23

L__I
F

where there are so few molecules that the electrons have a ve.-y gd chance
of reaching the anode without many coilietons The chance of broidonw

becomes less and additional voltage is needed to cause breakdown# a o

In Figure 3. It can be restated that as pressure decreases, the breakdown voltage decreases to a minimum at a critical pressure and thrm it increases until the value of breakdown In a perfect vac-um is readhed. Because of poostr ionization efficiency at very low pressures a long gap may break down at a lower voltage than a shorter gap. The dielectric saength of air at normal atmospherric temperature and pressure Is 22, 000 volts (rm) per inch, breakdown can occur at 341 volts at the cretical pressure. Breakdown votage decreases by approimatey 15 ercent for -rery 20 percmt deormse in pressure. The decrese in breakdown voitge of air with an increase in altitude makeS the testng of

airborne or-muulleborue equipment of-greater Importance because-spacings

which were mor -han adaquate at normal- asierclpressure may become

1hadequatatih altitudes. Time and-Bate of ApEgled Voltage

Test results are influenced-by the time-and rate of voltage aplcations In most otses, the breakdown voltage will tendto-Increae with Increasing rate

of voltage appleation

This resuts from the fact that both thermal-breakdown Also, the-- slower-the applicatonof

mechanism and-the discharge mechnism am-r- e-dUet althou in some cases the.disoare medhnism may ucasrapid falurebyprecling

crittcally high local iIr IntnsIes

voltage, the more likely fiiure will occur at a lower voltage Generally, voeltage Is Icreased sl y from zero to _te-requrevalue. unless breakdown occurs first or the voltige is rafsed Indiscrete steps. To t meaningul; -t

sttteussnt con.oeznl; dieletric strength about a particular 1!!atonali should spcfy or iM-tutime of voltage-appicatlon Just prior to breakdown StudiSes Indicate breakWown-vdte will tend todcesWfth uainot teat is iner.-I fftme 111 required for breakdowna to c-ccur. Bf-enof.thfi extemo B fied.ad s "e1)kaisstto lnt se Iti Instatiaos t- rloea eint

24

IE
W[E

Im

00

w
00

12ww0
oama

ON

IRS[t4

insulation Resistance
Measurement of insulation resistance is complicated by the time of electrification. Time of electrification means that as a potential difference Is a*ied to a spe imen, the current through It generaly decreases asymptotlally toward a limiting value which may be les than r percent of the current observed at the end ef the first minute. The decrease of -- rrent u is atributed to dielectric absorption changing capacitance, and 1he sweep of mobile ins to the electrode. The conventional arbitrary -time electrification of baa been snecifed -asone mnute. ~jvd of A2gied Voltane

Dtelectrtc Strena tesftS Thb - are-made to detect design deficiencies. flaws, damage In, manuktut, o my ignficat variation iun. dielectric, strengt The !evel of
vct ekiMatuld-give-consideratlon to both the inherent dielectric By ami hbrkdowavoltagells meaut theequivaleat-voltage at which:air Dielectric atregth tests which are nondeatructive are called "proof

brea~ownstrngthandthe spacing bre.kdown for the Insulatingwall thickness. apat _-*102ll breakdown~ Die
sc -brakao-

tric testing shimuld be grzht.etthan the gas ai

-voltage.

level and -lowerthan the delectric bre-aW

Th May cases-both volume and surface resistance or conductsace of thespecimen ma be voltage sensitive. in such a case the same voltage gr.adent must be used if results are to be consistent. A-tdlerance of 5 permt
aS ~ieall gien etwenap~ted and-specifted voltage. Insulation. resistance test anegmtrally perfobrmed at 5O90 volts Dc with special test s -s. GeticrellY,

insulation resistmw-e till decrease with an inerepae of -&*led voltae. The

variation in realiancee Is usually due to molsture or voids In Ite Insulationwa01 A tnddm chnagie in resistance can be a fobraewarnintg of an impending faiure.

S
IN

ISt2111501

-Humid-!y

..

Dilectric StrugMh
The relative humidity can greatly influence dielectic strength. When moisture is absorbed by or on the surface of the material tested the dielectric loss and surface conductivity is increased. The extent to which dielectric loss and surface conductivity are increased is dependent upon the nature *- the material being tested. However, some materials that absorb Mile or no mo-Istur may be affected because of the greatly Increased chemical effect. of discharge in the preence of moisture. Moisture lowers dielectc strengtb because water's dielectric properties are dominated by a polarization consiting of the orientation of the molecules by the action of the anplied field, the molecule having two hydrogen atoms with their +1 charges unsymmtrically disposed with respect to the oxygen atom with its -2 charge, giving it a permanent electric moment. Thlwpolarization accounts for the very high dielectric conftant of water at ordinary temperatures (tr= 80). The effects of relative humidity can be limited by standard conditioning p.--

-thereby

cedares

When moisture is-absorbed Into the pores of the inutlation, the

reistance bacomes-lower wMlthe power factor is increased. Volume re-

istivity-ls very smitive to-temperatre ch.gea_-

whle surface resistance

will d'ange rWty wit changesin-humiity. The change IS always exponentiaL If relative humi=ty is changed from 25 percent to 90 percent resistance can be chafted by a factor of cue million. When an insulang srface gets wet a thi film of water may be formed making the surface highly conductive.

severely affected at power freqaencle-(&-O-6O h-_). convrse mg be true. Capaci

rdvS-1

it has beer shown It Section hM that dielectric strength is not

At higher freuecies the ing the Insuatr a

e fly Icreae

Ii7= j

good conOucto-r where it ca--give orground c-ut electrical stgnds.

SA

Thickness The dielectric strength of electrical insulating material is dependent on specimen thickness. The dielectric strength for most materipls varies inversely as a fractional power of the specimen thickness. For some materials the dielertric strength varies as the reciprocal of the square root of the thickness. As a result, deletctrIc strength of an Insulating material Is equpJ t6 the breakdown voltage divided by the thickness and is usually espressed in volts per mil. Electrode Configuration Dielectric t~reth

In gener l, breakdown voltage tends to decrease with Increasing electrode ar and-this effect is more pronounced on thinner specimen& Results arc ilso affected by electr-ide material-since the-thermt and discharge mechanisms may be influenced by the thermal conductivi7 and work function of the electrode materi L The air breakdown voltage increases as spacing between the eec1 odes increases, but as spacing approaches zero distance the curve differs because of Praschens Law. After the critical gsp-spaclg is reached the -reakdown voltage of air varies Inversely proportional to the electrode epacing. According to Paschen's Law, the minimum vroltage required to breakdow air at ay sepa ation Is-equal to S-5 volts DC or AC peak. This is a significant point to be considered in insulation deiig and testin& For lower voltages, any separation between two wis is adeq'u.te as long as it is clean and dry.

Contour of Snecimen
The measured value of the insulation resistance of a specimen results from both its volume and surface -resistances. Because of the different prperlies in dIfferent materilas there is no assurance that, if material- A ha a.

higher insulatlon rtsimaice than material Bit will also have a higher resistnce
than B in the awlication for which it i intended. The contwur of the specimen can chbsge the L-1slaton resstance sign . ntl"

28

Environment
The eavironmeut can affect the heat transfer rate, external discharges, .dfield uniformity, thereby influencing test results. Results from one medium will differ when compared to those obtalned from a different medium, Most of today's equipment Is subject to a wide range of environmental conditions and exceeding high reliability requirements. Most equipment used in aircraft or missiles may be subjected to temperatures ranging from -70'F to several hundred degrees Fahrenheit under normal operating conditions. Vehicular vibration due tc, movement over roads or acceleradon In the air may cause considerable stress on equipment Handling subjects equipment to vaiovs degrees of wear and tear. Contmination Diei-ectri' Stre=wn Contamination of the surface of the Inuwlation with Arborne chemicals, dust, moisture, etc., can severely reduce the die!ectric strength of a material When impurities are present the amount and nature of the impurity, the materiel, size, shape and spacing of electrodes will all have a considerable affect upon the finale dielectric strength. Other contaminants such as grease and oil picked up during handling of the material may lower the dielectric strength. insu-atlon Resistance The resistance of clean, dry Insulation will tend to increase for hours when measured at a fixed voltage. On contaminated insulation, the stePd -value of resiatanc-. wll be reached quicker and usually at a much lower level Dirt or dust on an insulating surface will Increase the tendency for the formation of moisture films. A very thin moisture film may have such a high cor-uctivity as to reduce the Insulation resistance by a factor of several

thousand to one-

29

Waveshape Dielectric strength is influenced by the waveshape of the applied voltage. The peak value *1 a sine wave should be 1.414 times the rms voltage. Undersized or overloaded input circuits and excessive leakage current (low resistance) can cause waveshape distortion. Most distortion2 consists _f flattened, sharp, or jagged peaks higher than those found in undiato"-ed waveshapes. The maximum stress on - material being tested is due to the rms value when heating is most significant or pepk value of the voltage when highest voltage stress 1s most significant. The scale of a voltmeter is usually calibrated in terms of rms, but yields the true rms value only If the output is a good sinusoidal wave. If the waveshape is irregular, the peak. reading obtained may not be 1. 414 times the true rms voltage.

Aging is defined as any slow deterioration that ha6 not been satisfactorily explained. Mineral insulating materials like mica and quartz are slightly affected by aging, but most dielectrics are organic compounds and are liable to undergo changes. Aging consists of an oxidation of the material, reaction with the water molecules that penetrate its structure, the liberation of free ions, the breaking up of long chain molecules into shorter ones, the linking up of chain into a three-dimensional network or the breaking up of crystalline aggregates. Aging is accelerated by electrical use, chemical corrosion in the atmosphere, temperature changes, and exposure to light or ionized air. Aging is liable tc be Prcompanied by critical changes in dielectric properties which may lead to trouble where suitable insulation once existed. Previous Test Hletory DielectricStreh Every application of voltage to a dieltctr'c material deteriorates the insulatioa and lowers the dielectric strength. Deterioration can change the electrical parameters and physical characterlsticc. it is generally accepted that if a piece of maLerial can safely withstand a given voltage for one minute, it can withstand 80 percent of that voltage for hours. A one-minute test at a given voltage is cr nsidered equivalent to a five-secoad test at 120 percent of tha \oltage I As -i .electric m ,terial is continually retested, a lower test voltage give the same results as aid the initial teat voltage.

5can

I3

Insulation-Resistance As with dielectric strength, each application of voltage to an insulator deteriorates the iusulation and lowers the insulation resistance. An example of previous test history affect on insulation resistance would be aging. The longer a piece of electrical equipment is used, the more liable to breakdown it becomes.

WEI WEI

I 1

31

RIEFERENCES L. 2. Harold N. Mliller, Nodtrctive HihPtental Testin~- Hayden Book ("'nipany, Inc.,* New York 1884. Graham Lee Moses, Reuben Lee, and Robert ZK, Hillen, Lnsulration E~nerig undamental lake Publishing Company, Lake Forest, Illinois, 1958 "Electrical Insulating Materils", 1968 Book of ASTM Stadards. Part 29, American Society for Testing and Materials, Philadelphia, Pennsylvania. Harold N. Miller, "1AC and DC High Potential Testing", Electrical Pnineering, Vol 82, 1963. 1.- B. Kilman and J. P. Dallas, "High Potential Testing of Aircraft Equipment Electrical Insulation", Electrical Epg!Reerinz, Vol 75, p. 540-44.
R. J. Alke, ttVC Overpotential Testing on High Voltage Generators", Electrical Enalneerm , Vol 171, p. 1131.

3. 4. 5. 6.

7. 8& 9. 10.

Willis Jackson, The uation of Electrcal Egiment, John Wley and Sons, ipc., New Yoi*, 1954. Military Standard, ?ML-STD-202D, "Test Methods for Electronic and Electrical Component Parts", 14 April 1969. E. -L. Brancato, "Nondestructive Testing of Insulation", Electrical Engineeringa, Vol 72, p. 425. A. W. W. Cameron, "Nondestructive Tests for Generator Insulation", Electrical Eaneerin , Vol 71, p. 616.

BaMF11.

WUM W--12. 13.

Robert L Sarbar-her, SeD, Encyclpdic Dictionary of Electronicsan

Nucler Mlgneeri , Prentis-Hall, Inc., Englewood Cliffs, New Jersey, 1959. T. B. Owen, "Electrical System Transients and Sensitive Circuit Control", Electrical Englneerin, Vo-9 . 13027. Reference Data for Radio Eanglneers, Fourth Edition, International Telephone and Telegraph Corporation, New York, New York, 1967. 32

GWSSARY

Dielectric A nonconductIng sub, i ce or material through which, however, induction, magnetic lines of force, or electrostatic lines of force may pass. A dielectric Is a medium in which it is posqibile to produce and maintain

an electric field wi4lb litt-le or wo upply of energy from outside sources.

The energy required to produce the electric field Is recoverable, it; whole when the field is removed. In general, all insulating materials or In-.t, are dielectrics.

Dielectric Absorption
A phenomenan that occurs in imperfect dielectrics, whereby positive and negative charges are separated and then accumulated at certain regions

within the volume of the dielectric.

This phenomenon usually manifests

itself as a gradually decreasing current after the application of a fixed DC

I
-The

Dilectric Polarization
dipole moment per unit volume in a dielectric. It is a vector in the direction of the electric field and related to it by the following relationship:

P =D --cE= (2%co E
where P = Dieectric Polarization
D = Electric Dispiacement

(ec- c)E

c = Permittivity of the Dielectric C0 = Permittivity of Free Space Xe = Shsceptibility (th

o

- 1)

33

Db

mu

beie o h ielectric

stenttir-rctcamessthu~~ is-o the poetial-gadin coat which

Thpendo dieleticastrengtheo materal

teat.

An electrot Io a.dielectric body possess4ng separate electric poles of opposite sign and of a persnwent ornamipennaaent natur& It is an electrical anaog of a permanent muzet.

A detric In sidch a partof the we*3 rem.Av-4 jo-eatalitshan electric field i-te dielectric to izat returnedto the elecric mystem when to field to remvic The energy Wbah is not rntumad 6.z coverted Into hat in the dielectric. (AISE)

A material v4 such low conductivity t&at the flew of turreat trugh It under specific coneItonew m usuall, but.zst always, be negtcted& Insulator Stre~ The loading In pounds at which the Insulator tfil to perform ts function eltaer electrialy or mechanically, voltage and mechanIca tess being applied saimultanew-uuly. (AEE)

I
w

WS;_

The permittivity (e) of an Isotropic medium, for which the directions of the electric dispacement and the electric field intensity are the same at any point in the medium, Is the magnitude of the electric displacement density (I)at that point di'vied by the electric field Intensity thera The permittivity of a material is the value of the constant e appearing in~ the denominator of the Coulomb force equatio~n which expresses the force between two charges Immersed or Imbeddedi In the material. The permittivity is the relative permiilvity mutiplied by the permuittivity of free space, C=Kee~ where e permittivity

Ke =relative -permittivity (dielectric constant) c= perimittivity of free space. PbwerFactr ISectric rh elwhtia power systaim, the omine of the dtelectric phaxe &ngee or sine of Ike dielectric low angle.

35

FORMULAS

=C

11V

1 /2

6 C

ZAC

=D

~+i XC ZAc BDC

XC
Camdtive Reitance moac to Alternaflc Current
=
=

where:

Resiance to Direct Curranxt 100 MEGOHM 10-1 1

Let

RD C

=10 p

56

APPENDIX

WMen absorption losses to the capacitac, Cv, dielectric: Ke = Cr/O. are neet, a dielectric material can be represented by the followIng

Or, of a givena cotfigratlon of electrodes with a material as the dielectric,

The dielectric constant of a material Is the ratio of the capacitance,

WI
Figure 4

Phe Diaeof t A0e 8,Is heanguar difference betwee Ibe sinusoidal alterang PAtiu difference applied to a dielectric and the component of the r-n1tIC Altenatin curtreat having the ame period as the potential. diff-reame Dtelotric loaw Ansid, C Is the-difference between 90 degrees and the ~se cidctrc .nge. hey-oampoaent, wCV. Is affected by frequency and capaitacewhile th x-component Is afffeced by conductanc& If the capcitnceIsilw (ip) or fMeqenc is low (50-60 hi), the y-comtonent may swmler than the zecompnet. If capactance or frequency 15 high the ~be opps Is true sad is shown in Figure 5

MR-11i37

IMS W-:

WI

T a 2 cotaiua the rsult of c?~itive reactance, X, , Impance to alterWfng cuir-Mnt, ZAC, andi the ratio Xc/ZAC with the DC resistar.ce tqual to 10 megohra and c itance equal to 10 p7 for 101 60, 100, 1, 000 and 2, 000 hz = 100 ptf6r 5. 10, 60, 100 and 1,0 00 hz. The graph In iIgur

esows how XC,t-a

vartabe

ies with frequewy and capacitsce. The eowr$nate is

Cutoff values

ackc abscissa is-the varile frequey. the

for insulAtion re -stancean dielectric strength were made at the pot.s XC/ZAc ecusi to 10 percent and 90 percent This value ma be Incremsed or dcrmesd,

but 10 percent chsaws used ts an engineering awroximdation. In the area, beow 10 p eentdisletrt1c strepath would vrobtbl~y be tiw only test. Above
90 perceat insulation ristanc wodl probably by the only teat. The shaded areas show where both isuilation resistance and dielectric strength might be

applied.
A the point capac - 0 t

XC 0
2
=1-

5.ZAG=0-5 -1/4

?/ T(d

[uno02 + P

%SCW,=

3,XC

states thes current through the capacitor is 13T times the leakige rnlsftanc, Current.
In concluslen, the grep in Figure 6" sho- that when the kekae current Is 10 times the capwAtve cvrraxt oniy Inulwato reistafnce testing wou d probably be used and if the camaitive current l 10 times the leakage current -x-ly dielectric strengt tesrg would probably be used. All other times both

test might be apied.

KJ

IM

E38,

TABLE No. 2

z
.10

It

XC
1.59 x 109 2.65 X 10

Z.AC
1.6 X 09

C
0.99 -93

- 60
100 100o
2000 5* 10 o

2.84X 108
8 !.B9 x S10 8 1.03 x 10

1.59 X 10 1.59 X 107

6 x 10

.4 .15 os

3. 3.35 X 108 1.89 X10 1.03 X 108 1.o1 x idj

.2 X 100

.96

1.59 x

io8

60* 100*

2.65 X 107 1.59 X10?

.26 .16

-000*

.59 X106

,,,

.02

Slo00 pf

110

39

i It

i I i !

01

-'

I-

*1b

rw
I I
CO V

jI
040